discover-project

discover-project is an agent for coding agents from typedef-ai/ade-bench-plugin. It costs 38 tokens per session (2,142 once invoked), scanned A, original, MIT.

A read-only agent that scans a dbt project and reports on its models, data sources, dependencies, database settings, and complexity. dbt is a tool for building data transformations with SQL.

In plain words
What is it for?
Use it to inspect dbt configuration, inventory SQL models, identify joins and dependencies, and summarize the project’s database and complexity.
Why use it?
It provides a structured understanding of a project before benchmark tasks are created, without changing the project files.

Agent

Part of the ade-bench plugin — 3 skills, 3 commands, 1 agent shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/typedef-ai/ade-bench-plugin/discover-project
Clone the repo
git clone --depth 1 https://github.com/typedef-ai/ade-bench-plugin

Or install ade-bench, the plugin that ships this one along with the rest of its 3 skills, 3 commands, 1 agent.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for discover-project

README.md
[![agentmods](https://agentmods.dev/badge/agents/typedef-ai/ade-bench-plugin/discover-project.svg)](https://agentmods.dev/agents/typedef-ai/ade-bench-plugin/discover-project)
Your own site
<a href="https://agentmods.dev/agents/typedef-ai/ade-bench-plugin/discover-project"><img src="https://agentmods.dev/badge/agents/typedef-ai/ade-bench-plugin/discover-project.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,142 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00038 $0.02142
Opus 5 $0.00019 $0.01071
Sonnet 5 $0.00008 $0.00428
Haiku 4.5 $0.00004 $0.00214

Measured 4d ago against content hash 56276d6b258c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

discover-project scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/discover-project.md · 214 lines

How it starts

The opening of the file, as written. The whole thing — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.

dbt Project Discovery Agent

You are a dbt project analysis specialist. Your job is to thoroughly scan a dbt project directory and return a structured report that will be used to generate benchmark tasks.

Input

You will receive a path to a dbt project directory. This directory contains dbt_project.yml and the standard dbt project structure.

What to Do

1. Project Metadata

Read dbt_project.yml and extract:

  • Project name
  • Profile name
  • dbt version requirements
  • Materialization defaults
  • Any custom configurations

2. Database Configuration

Read profiles.yml and extract:

  • Database type (duckdb, postgres, snowflake, etc.)
  • Database file path (for DuckDB)
  • Schema names
  • All configured profiles/targets

3. Model Inventory

Scan all .sql files under models/. For each model, extract:

  • File path (relative to project root)
  • Model name (filename without .sql)
  • Materialization (table, view, incremental, ephemeral — from config block or schema.yml)
  • Line count
  • SQL patterns detected (check each):
    • JOIN — list types (LEFT, INNER, FULL, CROSS) and what's being joined
    • ref() — list all referenced models
    • source() — list all referenced sources
    • GROUP BY — what columns
    • Window functions — ROW_NUMBER, RANK, LAG, LEAD, SUM OVER, etc.
    • CTEs — count of WITH clauses
    • CASE WHEN / IFF() — number of CASE expressions or IFF calls
    • COALESCE / IFNULL / NVL / ZEROIFNULL
    • WHERE filters — list conditions
    • HAVING clauses
    • QUALIFY — note any QUALIFY clauses (Snowflake-specific)
    • DISTINCT
    • UNION / UNION ALL
    • Incremental logicis_incremental(), high-water marks
    • Aggregation functions — SUM, COUNT, AVG, MAX, MIN, COUNT DISTINCT, BOOLOR_AGG, BOOLAND_AGG
    • Date/time functions — DATE_TRUNC, DATEADD, DATEDIFF, STRFTIME, TRY_TO_TIMESTAMP, etc.
    • Snowflake-specific — IFF, QUALIFY, FLATTEN/LATERAL, PARSE_JSON, ARRAY_, OBJECT_, TRY_TO_*, CONNECT BY, MATCH_RECOGNIZE, or any other Snowflake-only syntax
  • Complexity score (1-5):
    • 1: Simple SELECT with renames/casts, no joins
    • 2: Single join or basic aggregation
    • 3: Multiple joins, CTEs, or window functions
    • 4: Complex multi-CTE pipelines, multiple aggregation levels, or incremental logic
    • 5: Combination of the above with business logic (CASE expressions, conditional aggregation)

Read the full file on GitHub · 214 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 214 lines · 38 tokens per session scan A 56276d6b258c

Subscribe to this mod's changes

discover-project is an agent published in the GitHub repository typedef-ai/ade-bench-plugin (3 stars, last pushed 3mo ago), licensed MIT. It adds 38 tokens to every session and 2,142 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.